Threshold optimization and random undersampling for imbalanced credit card data

Journal of Big Data

Table 18 Results for the extremely randomized trees depth 8 classifier

Technique	Threshold	TPR	FPR	FNR	TNR	F-meas	G-mean	MCC	Precision
F-meas	0.2232	0.9423	0.1047	0.0577	0.8953	0.0308	0.9183	0.1139	0.0157
F-meas NC	0.2306	0.9384	0.0969	0.0616	0.9031	0.0339	0.9204	0.1196	0.0173
G-mean	0.2165	0.9459	0.1137	0.0541	0.8863	0.0286	0.9155	0.1092	0.0145
G-mean NC	0.2319	0.9380	0.0951	0.0620	0.9049	0.0344	0.9211	0.1205	0.0175
MCC	0.2148	0.9465	0.1165	0.0535	0.8835	0.0280	0.9143	0.1079	0.0142
MCC NC	0.2372	0.9364	0.0905	0.0636	0.9095	0.0368	0.9226	0.1247	0.0188
Precision	0.2275	0.9411	0.0987	0.0589	0.9013	0.0324	0.9208	0.1172	0.0165
Precision NC	0.3903	0.8921	0.0218	0.1079	0.9782	0.1546	0.9339	0.2653	0.0876
C	0.5000	0.8665	0.0073	0.1335	0.9927	0.2946	0.9273	0.3894	0.1794
D	0.5000	0.8665	0.0073	0.1335	0.9927	0.2946	0.9273	0.3894	0.1794